ABSTRACT It was originally thought that all immune cells were derived from a single hematopoietic stem cell (HSC). However, it is now clear that the mouse B1a B cell subset and the follicular (B2) B cell subset are derived from different HSCs that reside in fetal liver (FL) and adult bone marrow (BM), respectively. There is also mounting evidence that B cells with the cell surface phenotype of B1b and marginal zone (MZ) B cell subsets can be derived from both FL and adult BM in mice, but whether the B cells derived from these different sources have distinct properties is unclear. Whether this situation is recapitulated in humans is not known. However, we have shown that total CD34+ HSCs mobilized from adult human BM cannot effectively reconstitute a B cell compartment when transplanted into severely immune deficient (NOD/SCID/common cytokine receptor ? chain knockout; NSG) mice. In contrast, FL and umbilical cord blood (UCB) CD34+ cells are efficient in this regard but give rise to a B cell compartment that phenotypically and functionally resembles that of human newborns and young children. The Lin28b miRNA binding protein has been shown to play a major role in regulating the differentiation potential of the mouse HSCs that give rise to B1a B cells and other ?innate-like? lymphoid subsets (the ?fetal/neonatal? lineages). Strikingly, when HSCs from adult mouse BM are transduced to express Lin28b they are licensed to give rise to B1a B cells and other fetal/neonatal lymphoid lineages in myeloablated host mice. A potential role for multiple HSC types and Lin28b regulation of HSC differentiation potential has not been investigated in humans. However, Lin28b is expressed in human FL and UCB HSCs but is not detectable in human BM. Moreover, patients that undergo transplantation with HSCs derived from mobilized BM HSCs after myeloablation therapy do not mount robust antibody responses to many antigens and are very susceptible to viral and bacterial infections. Certain antibody responses in mice are predominantly produced by the fetal/neonatal B cell subsets. In total, these data suggest the hypothesis that adult human BM HSCs lack the ability to reconstitute the fetal/neonatal B cell lineages due to absence of the appropriate HSC subset. Moreover, we suggest that, as is the case in mice, adult BM HSCs can be licensed to reconstitute the fetal/neonatal B cell lineages via enforced expression of Lin28b.